Chris Wright
@kernelcdub
Red Hat CTO. Tezos Foundation council member. Passion for open source SW innovation. Father and husband. Cyclist. Human.
ID: 583450546
18-05-2012 04:09:51
1,1K Tweet
6,6K Followers
265 Following
Here's how we're achieving R1 like reasoning with small models leveraging probabalistic inference-time scaling w/out using DeepSeek or derivatives. Results are impressive! MATH w/ Llama 8B approaches GPT-4o, and w/ Qwen2.5 Math 7B Instruct hits o1 level. red-hat-ai-innovation-team.github.io/posts/r1-like-…